ABSTRACT
Objective:To explore the effect of joint segmentation model of myocardial-fibrotic region based on deep learning in quantitative analysis of myocardial fibrosis in patients with dilated cardiomyopathy(DCM).Methods:The data of 200 patients with confirmed DCM and myocardial fibrosis in the left ventricle detected by cardiac MR-late gadolinium enhancement (CMR-LGE) in Xuzhou Central Hospital from January 2015 to April 2022 were retrospectively analyzed. Using a complete randomized design, the patients were divided into training set ( n=120), validation set ( n=30) and test set ( n=50). The left ventricle myocardium was outlined and the normal myocardial region was selected by radiologists. Fibrotic myocardium was extracted through calculating the threshold with standard deviation (SD) as a reference standard for left ventricle segmentation and fibrosis quantification. The left ventricular myocardium was segmented by convex prior U-Net network. Then the normal myocardial image block was recognized by VGG image classification network, and the fibrosis myocardium was extracted by SD threshold. The myocardial segmentation effect was evaluated using precision, recall, intersection over union (IOU) and Dice coefficient. The consistency of myocardial fibrosis ratio in left ventricle obtained by joint segmentation model and manual extraction was evaluated with intra-class correlation coefficient (ICC). According to the median of fibrosis rate, the samples were divided into mild and severe fibrosis, and the quantitative effect of fibrosis was compared by Mann-Whitney U test. Results:In the test set, the precision of myocardial segmentation was 0.827 (0.799, 0.854), the recall was 0.849 (0.822, 0.876), the IOU was 0.788 (0.760, 0.816), and the Dice coefficient was 0.832 (0.807, 0.857). The consistency of fibrosis ratio between joint segmentation model and manual extraction was high (ICC=0.991, P<0.001). No statistically significant difference was found in the ratio error between mild and severe fibrosis ( P>0.05). Conclusions:The joint segmentation model realizes the automatic calculation of myocardial fibrosis ratio in left ventricle, which is highly consistent with the results of manual extraction. Therefore, it can accurately realize the automatic quantitative analysis of myocardial fibrosis in patients with dilated cardiomyopathy.
ABSTRACT
Objective To investigate the clinical value of 2 dimension late Gadolinium enhancement MRI (LGE-MRI) technique for the evaluation of atrial myocardial fibrosis in patients with atrial fibrillation. Methods Forty-nine cases of atrial fibrillation in our hospital from March 2015 to December 2016 were retrospectively collected. The LGE-MR was acquired by the Siemens 3.0 T MR machine before the catheter ablation.The findings of LGE-MR were evaluated by two experienced doctors. The left atrium(LA)were manually segmented into 8 regions in axial view.All patients were classified into 4 stages based on the extent of enhancement, stage 0: absence of enhancement, stage Ⅰ: enhancement appeared in minimal two consecutive slices in single region,stageⅡ:enhancement in two regions,stageⅢ:enhancement in three or more regions. All electroanatomic maps were obtained after electrical conversion during catheter ablation. The Kappa test was used to assess the consistency of LGE-MRI left atrial myocardial fibrosis and CARTO system of the left atrial endocardial voltage reconstruction. Results Forty-nine cases of atrial fibrillation with LGE-MRI and CARTO were included. There were 17 cases of atrial fibrosis stage 0,10 cases of stageⅠ,11 cases of stageⅡ,11 cases of stageⅢaccording to LGE-MRI findings;There were 17 cases of atrial fibrosis stage 0,19 cases of stageⅠ,12 cases of stageⅡ,11 cases of stage Ⅲ with reference to CARTO findings. The diagnostic accuracy of the LGE-MRI atrial fibrosis was 81.6%(40/49),of which the correlation was good(Kappa= 0.751,P<0.001). Conclusions LGE-MRI can accurately assess the degree of left atrial myocardial fibrosis in patients with atrial fibrillation,help to select the proper candidate and strategy in catheter ablation.